CN114341366A

CN114341366A - Biomarkers and methods for personalized treatment of small cell lung cancer using KDM1A inhibitors

Info

Publication number: CN114341366A
Application number: CN201980100040.4A
Authority: CN
Inventors: F·西赛里; N·萨基洛托; S·鲁纳迪
Original assignee: Oryzon Genomics SA
Current assignee: Oryzon Genomics SA
Priority date: 2019-07-05
Filing date: 2019-07-05
Publication date: 2022-04-12
Also published as: JP2022546908A; WO2021004610A1; EP3994280A1

Abstract

Biomarkers and methods for predicting responsiveness of a patient with Small Cell Lung Cancer (SCLC) to treatment with a KDM1A inhibitor are disclosed, as well as methods of treating a subset of SCLC patients identified using the methods.

Description

Biomarkers and methods for personalized treatment of small cell lung cancer using KDM1A inhibitors

Technical Field

The present invention relates to biomarkers and methods for the personalized treatment of Small Cell Lung Cancer (SCLC) using KDM1A inhibitors. The present invention provides methods of identifying SCLC patients who may benefit from treatment with a KDM1A inhibitor and methods of treating such patients with a KDM1A inhibitor.

Background

Lysine-specific demethylase 1(LSD1, also known as KDM1A) is a histone modifying enzyme responsible for demethylation of dimethyl histone 3 lysine 4(H3K4me2) (Shi et al, Cell 2004). KDM1A overexpression has been associated with disease progression and poor prognosis in several human cancers, and its inhibition has been shown to reduce growth, migration and invasion of cancer cells. Therefore, KDM1A has been considered as a target for the development of new drugs for the treatment of cancer, and several KDM1A inhibitors are currently in clinical trials in oncology.

In particular, KDM1A inhibitors have been reported to be effective in treating Small Cell Lung Cancer (SCLC). It has been shown that KDM1A inhibition reduces the in vitro proliferation of SCLC Cell lines and delays tumor growth in SCLC-bearing xenograft mice (Mohammad et al 2015, Cancer Cell 28, 57-69). However, published data suggests that while certain SCLCs are highly sensitive to KDM1A inhibition, sensitivity to KDM1A inhibition is not a universal feature of SCLC cells. There is a need to develop methods for personalized SCLC treatment using KDM1A inhibitors, and in particular, to develop patient selection methods to identify those SCLC patients that would benefit from or most benefit from treatment with KDM1A inhibitors.

It is therefore the technical problem of the present invention to provide means and methods for identifying and treating SCLC patients best suited for treatment with KDM1A inhibitors.

Disclosure of Invention

The present invention provides tools and methods for the personalized treatment of Small Cell Lung Cancer (SCLC) using KDM1A inhibitors.

In one aspect, the invention provides a method of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor.

For example, patients are identified as more likely to respond to treatment comprising a KDM1A inhibitor than patients with SCLC and having levels of ASCL1 and SOX2 measured from the patient's sample prior to initiating treatment comprising a KDM1A inhibitor (when the level of each of ASCL1 and SOX2 in the sample does not exceed a threshold). Conversely, the latter patient (the level of each of ASCL1 and SOX2 in the sample not exceeding the threshold) is unlikely to respond to a treatment comprising a KDM1A inhibitor. This exemplary interpretation applies to all aspects and uses provided herein that relate to, encompass or include identifying SCLC patients that are more likely to respond/respond to treatments comprising KDM1A inhibitors and the like.

In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In another aspect, the invention provides a method of identifying SCLC patients who may benefit from treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as a patient that may benefit from treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein when the score in the sample exceeds a threshold, the patient is identified as a patient that may benefit from treatment comprising a KDM1A inhibitor.

In a further aspect, the invention provides a method of predicting the responsiveness of an SCLC patient to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to the commencement of the treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to be responsive to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the present invention provides a method of assessing the likelihood of a SCLC patient responding to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the present invention provides a method of assessing the likelihood of a SCLC patient responding to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from a SCLC patient prior to the commencement of a treatment comprising a KDM1A inhibitor. In some embodiments, SCLC is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein SCLC is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the invention provides a method of selecting a treatment for an SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating the treatment. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the method comprises providing a recommendation that the selected treatment for the patient comprise a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, and providing a recommendation that the selected treatment for the patient comprise a KDM1A inhibitor when the score in the sample exceeds a threshold. In other words, for example, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, or when the score in the sample exceeds a threshold, the treatment selected for the SCLC patient is a treatment comprising a KDM1A inhibitor. When the level of each of ASCL1 and SOX2 does not exceed a threshold, or when the score does not exceed a threshold, the patient may be considered other treatment options than treatment comprising a KDM1A inhibitor, such as treatment with a drug/therapeutic agent other than a KDM1A inhibitor.

In a further aspect, the invention provides a method of treating a SCLC patient comprising administering to the patient a therapeutically effective amount of a treatment comprising a KDM1A inhibitor if the patient has been identified as more likely to respond to treatment comprising a KDM1A inhibitor using a method according to any preceding aspect before commencing treatment comprising a KDM1A inhibitor.

In a further aspect, the invention features a method of treating a SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment, the patient identified as more likely to respond to treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, and administering a therapeutically effective amount of treatment comprising a KDM1A inhibitor to the patient if the patient is identified as more likely to respond.

In a further aspect, the invention features a method of treating a SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment, using these levels to generate a score for the sample, the patient identified as more likely to respond to treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold, and administering a therapeutically effective amount of treatment comprising a KDM1A inhibitor to the patient if the patient is identified as more likely to respond.

In a further aspect, the invention provides a KDM1A inhibitor for use in treating an SCLC patient, wherein the patient has been identified as more likely to respond to treatment comprising a KDM1A inhibitor using a method according to any preceding aspect prior to initiating treatment comprising a KDM1A inhibitor.

In a further aspect, the invention provides the use of ASCL1 and SOX2 in a method of identifying SCLC patients more likely to respond to a treatment comprising a KDM1A inhibitor.

In a further aspect, the invention provides the use of ASCL1 and SOX2 in a method of assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor.

In a further aspect, the invention provides the use of one or more agents for measuring the levels of ASCL1 and SOX2 in a method of identifying SCLC patients more likely to respond to a treatment comprising a KDM1A inhibitor.

In a further aspect, the present invention provides the use of one or more agents for measuring the level of ASCL1 and SOX2 in a method of assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor.

In a further aspect, the invention provides the use of ASCL1 and SOX2 for the preparation of a diagnostic for identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor.

In a further aspect, the present invention provides the use of ASCL1 and SOX2 for the preparation of a diagnostic for assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor.

In another aspect, the invention provides a kit for identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, the kit comprising one or more reagents for measuring the levels of ASCL1 and SOX2 in a sample, and optionally, instructions for use.

In another aspect, the invention provides a kit for assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor, the kit comprising one or more reagents for measuring the level of ASCL1 and SOX2 in a sample, and optionally, instructions for use.

Drawings

FIG. 1:the dot plots represent the expression (Log2(RPKM) values) of ASCL1 (Y-axis) and SOX2 (X-axis) of SCLC cell lines sensitive (grey dots), partially sensitive (open squares) or resistant (black triangles) to KDM1A inhibition as measured by RNA-seq, as described in more detail in example 2.

FIG. 2:the dot plots represent gene expression (absolute Cp values) of ASCL1 and SOX2 of SCLC cell lines sensitive (grey dots), partially sensitive (open squares) or resistant (black diamonds) to KDM1A inhibition as measured by qRT-PCR, as performedAs described in example 3. The plotted values are the average of independent experiments. The values in parentheses have Cp values expressed by SOX2 of 40 or more.

FIG. 3:the dot plots represent gene expression (RMA values) of ASCL1 and SOX2 in an expanded set of SCLC cell lines sensitive (grey dots), partially sensitive (open squares) or resistant (black triangles) to KDM1A inhibition as measured by microarray Affymetrix analysis, as described in example 4.

FIG. 4:ROC curves based on gene expression of ASCL1 (fig. 4A) and SOX2 (fig. 4B) were used to distinguish KDM1Ai sensitive and resistant SCLC cells as described in example 4. The sensitivity and specificity for a given threshold and their respective confidence intervals are shown in the table below each graph.

FIG. 5:western Blot (WB) of ASCL1 and SOX2 protein levels (FIG. 5A) and quantification (FIG. 5B) in NCI-H146, NCI-H510A, NCI-H446 and NCI-H526 cell lines as described in example 5.1.

FIG. 6:correlation between protein and mrna (ccle affymetrix) levels of SOX2 (fig. 6A) and ASCL1 (fig. 6B), as described in example 5.1.

FIG. 7:fluorescent immunohistochemical staining according to SOX2 (fig. 7A), ASCL1 (fig. 7B) and the negative control (no primary antibody, fig. 7C) of example 5.2 showed low/undetectable levels of the two biomarkers in NCI-H526 cells, low/undetectable levels of ASCL1 in NCI-H446 cells, and the highest level of the two biomarkers in NCI-H146 cells, consistent with their mRNA levels. DAPI (4', 6-diamidino-2-phenylindole) is a fluorescent dye that binds strongly to the A-T rich region of DNA and stains the nucleus. No signal was detected in the negative control containing secondary antibody alone (AF 546: Alexa Fluor 546).

FIG. 8:correlation between protein and mrna (ccle affymetrix) levels of SOX2 (fig. 8A) and ASCL1 (fig. 8B), as described in more detail in example 5.2.

FIG. 9:representative ASCL1, SOX2 from SCLC PDX TMA and their corresponding DAPI fluoroimmune panels for each

stain classification level

0, 1, 2 and 3 are shownImages were visualized as described in example 5.3.

FIG. 10:RNASeq (Log) of SOX2 (FIGS. 10A and 10B) and ASCL1 (FIGS. 10C and 10D) for two independent experiments are shown₂FPKM) and IF (visual score) with a confidence interval of 95% (coefficients are assigned at the bottom of each graph), as described in example 5.3.

FIG. 11:WB and Ponceau staining of ASCL1 and SOX2 in the exosome fraction (Ponceau staining) according to example 6. Consistent with mRNA expression, ASCL1 was not detected in the exosomes derived from NCI-H446 and NCI-H526, and SOX2 was not present in the exosomes derived from NCI-H526.

FIG. 12:according to example 6, WB (left) and ponceau red staining of ASCL1, SOX2 and CD151 (lung cancer specific exosome markers) in exosomes and corresponding parental cells (right). The ASCL1, SOX2 and CD151 signals in the exosome fraction were significantly reduced or eliminated 48 hours after treating NCI-H510A cells with 5 μ M GW4869 (an exosome production inhibitor), while the expression of these proteins in the cells was unaffected, indicating that detection of ASCL1, SOX2 and CD151 is exosome-specific.

Detailed Description

Definition of

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below.

All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety.

Unless otherwise indicated, the nomenclature used in this application is based on the IUPAC systematic nomenclature.

Unless otherwise indicated, any open valency appearing on a carbon, oxygen, sulfur or nitrogen atom in the structures herein indicates the presence of hydrogen.

The stereochemical definitions and conventions used herein generally follow the definitions of S.P. Parker, eds, McGraw-Hill Dictionary of Chemical Terms (1984) McGraw-Hill Book Company, New York; and Eliel, E. and Wilen, S., "Stereochemistry of Organic Compounds", John Wiley & Sons, Inc., New York, 1994. In describing optically active compounds, the prefixes D and L, or R and S, are used to denote the absolute configuration of the molecule with respect to its chiral center. The substituents attached to the chiral center of interest are arranged in the order Cahn, Ingold, and Prelog (Cahn et al Angew. chem. Inter. ed., 1966,5, 385; errata 511). The prefixes D and L or (+) and (-) are employed to denote the sign of rotation of the compound to plane polarized light, where (-) or L denotes that the compound is left-handed. Compounds with the prefix (+) or D are dextrorotatory.

The terms "optional" or "optionally" mean that the subsequently described event or circumstance may, but need not, occur, and that the description includes instances where the event or circumstance occurs and instances where it does not.

The term "pharmaceutically acceptable salt" refers to salts that are not biologically or otherwise undesirable. Pharmaceutically acceptable salts include acid addition salts and base addition salts. Pharmaceutically acceptable salts are well known in the art.

The term "pharmaceutically acceptable acid addition salts" denotes those pharmaceutically acceptable salts formed with inorganic acids such as hydrochloric acid, hydrobromic acid, sulfuric acid, nitric acid, carbonic acid, phosphoric acid and organic acids selected from the group consisting of aliphatic, alicyclic, aromatic, araliphatic, heterocyclic, carboxylic and sulfonic organic acids, for example formic acid, acetic acid, propionic acid, glycolic acid, gluconic acid, lactic acid, pyruvic acid, oxalic acid, malic acid, maleic acid, malonic acid, succinic acid, fumaric acid, tartaric acid, citric acid, aspartic acid, ascorbic acid, glutamic acid, anthranilic acid, benzoic acid, cinnamic acid, mandelic acid, pamoic acid, phenylacetic acid, methanesulfonic acid, ethanesulfonic acid, p-toluenesulfonic acid and salicylic acid.

The term "pharmaceutically acceptable base addition salts" denotes those pharmaceutically acceptable salts formed with organic or inorganic bases. Examples of acceptable inorganic bases include sodium, potassium, ammonium, calcium, magnesium, iron, zinc, copper, manganese, and aluminum salts. Salts derived from pharmaceutically acceptable organic non-toxic bases include salts of primary, secondary and tertiary amines, substituted amines (including naturally occurring substituted amines), cyclic amines and basic ion exchange resins, such as isopropylamine, trimethylamine, diethylamine, triethylamine, tripropylamine, ethanolamine, 2-diethylaminoethanol, trimethylamine, dicyclohexylamine, lysine, arginine, histidine, caffeine, procaine, hydrabamine, choline, betaine, ethylenediamine, glucosamine, methylglucamine, theobromine, purine, piperazine, piperidine, N-ethylpiperidine and polyamine resins.

The terms "KDM 1A inhibitor" or "KDM 1 Ai" are used interchangeably and as used herein refer to any compound capable of inhibiting KDM1A activity. Methods for determining the inhibitory activity of KDM1A are well known in the art. In preferred embodiments, the KDM1A inhibitor is a small molecule. Examples of KDM1A inhibitors are described in more detail elsewhere herein.

As used herein, "small molecule" refers to an organic compound having a molecular weight of less than 900 daltons, preferably less than 500 daltons. Molecular weight is the mass of a molecule, calculated as the sum of the atomic weights of each constituent element multiplied by the atomic number of that element in the formula.

By "treatment comprising a KDM1A inhibitor" is meant any therapy or treatment regimen that incorporates a KDM1A inhibitor, whether as the sole Active Pharmaceutical Ingredient (API) or in combination with one or more other APIs (such as other anti-cancer agents). Such treatments comprising KDM1A inhibitors are typically in the form of pharmaceutical compositions. If a treatment comprising a KDM1A inhibitor comprises one or more APIs in addition to a KDM1A inhibitor, it may be administered in the form of a single pharmaceutical composition incorporating all of the APIs, or may be administered in the form of separate pharmaceutical compositions for each API (i.e., KDM1A inhibitor and one or more other APIs), may be administered by the same or different routes (e.g., may be administered orally in one way, parenterally in another), and may be administered simultaneously or sequentially.

The terms "pharmaceutical composition" and "pharmaceutical formulation" are used interchangeably to refer to a composition (e.g., a mixture or solution) comprising a therapeutically effective amount of an active pharmaceutical ingredient (e.g., a KDM1A inhibitor) and one or more pharmaceutically acceptable excipients to be administered to a mammal, such as a human in need thereof.

The term "pharmaceutically acceptable" refers to the properties of a material that can be used to prepare a pharmaceutical composition that is generally safe, non-toxic, neither biologically nor otherwise undesirable, and acceptable for veterinary as well as human pharmaceutical use.

The terms "pharmaceutically acceptable excipient", "pharmaceutically acceptable carrier" and "therapeutically inert excipient" are used interchangeably to refer to any pharmaceutically acceptable ingredient in a pharmaceutical composition that is not therapeutically active and is non-toxic to the subject to which it is administered, such as disintegrants, binders, fillers, solvents, buffers, tonicity agents, stabilizers, antioxidants, surfactants, carriers, diluents or lubricants used in formulating pharmaceutical products.

The term "therapeutically effective amount" (or "effective amount") means an amount of a compound of the present invention that, when administered to a patient, (i) treats or prevents a particular disease, (ii) attenuates, ameliorates, or eliminates one or more symptoms of a disease, or (iii) prevents or delays the onset of one or more symptoms of a disease. The therapeutically effective amount will vary depending upon the compound, the condition being treated, the severity of the condition being treated, the age and relative health of the patient, the route and form of administration, the judgment of the attending physician or veterinary practitioner, and other factors.

The term "treatment" of a disease (e.g., SCLC), as well as various parts-of-speech forms thereof, includes reversing, alleviating, or inhibiting the progression of the disease or one or more symptoms thereof.

"patient" or "subject" are used interchangeably and refer to a mammal in need of treatment. Mammals include, but are not limited to, primates (e.g., humans and non-human primates, such as monkeys), domesticated animals (e.g., cows, sheep, cats, dogs, and horses), and laboratory animals (mice, rats, guinea pigs, and the like). In a preferred embodiment, the patient is a human. Any subject involved in a clinical study trial is intended to be included as a patient. The patient may have been previously treated with, for example, other drugs and/or any KDM1A inhibitor. In one aspect, the patient has not been previously treated with any KDM1A inhibitor. The patient may be being treated with other drugs, particularly when obtaining a sample, however, the patient should not be treated with any KDM1A inhibitor when obtaining a sample for use in a method according to the invention (i.e. the patient should not be treated with a KDM1A inhibitor at the same time when obtaining a sample). Alternatively, if the biomarker level is still likely to be modulated by the (remaining) KDM1A inhibitor within the time period (e.g. if the biomarker level has not returned to the level prior to previous treatment with (or administration of) a KDM1A inhibitor), then the patient should not be being treated with any KDM1A inhibitor for the time period prior to obtaining the sample. For example, within two weeks, or more preferably within one month, prior to obtaining the sample, the patient should not be being treated with any KDM1A inhibitor. The latter is to avoid that biomarker levels are still regulated by the (remaining) KDM1A inhibitor.

As used herein, the term "biomarker" or "marker" refers to a protein or polynucleotide, the expression or presence of which in or on mammalian tissue or cells can be detected by standard methods (or methods disclosed herein), and correlates with the sensitivity of mammalian cells or tissues to a treatment comprising a KDM1A inhibitor. Biomarkers according to the invention are ASCL1 and SOX 2.

As used herein, the term "measuring" a level of a biomarker refers to experimentally determining the amount of the biomarker in a sample using an appropriate detection method as described elsewhere herein.

As used herein, the term "threshold" refers to a predetermined value, line, or more complex n-dimensional function that defines a boundary between two classes/subsets of a population, e.g., SCLC patients that are more likely to respond to KDM1A inhibitor therapy relative to SCLC patients that are less likely to respond to KDM1A inhibitor therapy. In the method according to the invention, different thresholds may be applied to the levels of individual biomarkers (i.e. the biomarkers ASCL1 and SOX2 each have their respective threshold) or to scores derived from the biomarker levels by using a classification algorithm as described elsewhere herein. As will be understood by those skilled in the art, a threshold is established to best distinguish between different classes of samples. The threshold may be established according to methods known in the art. In general, thresholds can be determined experimentally or theoretically using training set samples with known sensitivity or resistance to KDM1A inhibitor therapy. The training sample may be, for example, an SCLC cell line, a patient-derived xenograft (PDX), or a human clinical sample with known sensitivity or resistance to KDM1A inhibitor therapy. As one of ordinary skill in the art will recognize, the threshold may also be arbitrarily selected based on existing experimental and/or clinical and/or regulatory requirements. Preferably, thresholds are established in order to obtain optimal sensitivity and specificity as a function of the test and the benefit/risk balance (clinical outcome of false positives and false negatives). In general, optimal sensitivity and specificity (and thresholds) can be determined using Receiver Operating Characteristic (ROC) curves based on experimental data, as shown in the accompanying examples. In some embodiments, the threshold is a threshold value. In some embodiments, the cutoff values for the biomarkers are derived from ASCL1 and SOX2 levels measured in one or more samples in patient-derived SCLC cells that are sensitive or resistant to treatment comprising a KDM1A inhibitor. In some embodiments, the cut-off values are derived from ASCL1 and SOX2 levels measured in one or more samples of a (human) patient who is responsive or non-responsive to a treatment comprising a KDM1A inhibitor. In some embodiments, the cutoff values are derived from levels of ASCL1 and SOX2 measured in one or more samples obtained from a patient-derived xenograft model that is responsive or non-responsive to a treatment comprising a KDM1A inhibitor. In some embodiments, the cut-off value is obtained from the mRNA level of the biomarker. In some embodiments, the cutoff value is obtained from the protein level of the biomarker.

As used herein, the term "score" refers to the output calculated by a classification algorithm from the measured biomarker levels in a sample. The score will be compared to/with a threshold and used to determine whether the patient from whom the sample was derived is more or less likely to respond to a treatment comprising a KDM1A inhibitor. For example, when the score exceeds a threshold, the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor/the patient is identified as he/she may benefit from treatment comprising a KDM1A inhibitor.

As used herein, a "classification algorithm" is a mathematical function that is used to calculate a score for a sample and to evaluate ("classify") to which class the sample belongs, i.e., whether it exceeds a threshold. Classification algorithms are well known in the art. Examples of classification algorithms include: linear classifiers, Fisher linear discriminant, linear Boolean (Boolean) classification, logistic regression, naive bayes classifier, perceptron, support vector machine, least squares support vector machine, quadratic classifier, kernel estimation, k-nearest neighbor algorithm, decision tree, random forest, neural network, learning vector quantization. Software packages incorporating one or more classification algorithms are readily available or downloaded online, including, for example, DTREG, XLSTAT, http:// www.support-vector-machines. In some embodiments, the classification algorithm is a boolean function (truth function).

The samples may be classified using a Boolean conjunction function (Boolean conjunction function) a AND B, where a AND B evaluate whether the level of each of the biomarkers (ASCL1 AND SOX2) in the sample is above the respective threshold for that biomarker. The boolean conjunction function gives a score that is typically represented by 1 (true value) when all criteria are met (e.g., if the respective levels of the biomarkers (ASCL1 and SOX2) in the sample are above/exceed the respective thresholds for the biomarkers), or by 0 (false value) when one (or both) of the criteria are not met (e.g., if the level of only one of the biomarkers (ASCL1 and SOX2) in the sample is above/exceeds the respective thresholds for the biomarkers, or the level of no biomarker in the biomarkers (ASCL1 and SOX2) is above/exceeds the respective thresholds for the biomarkers). The threshold applied to the score generated by the boolean algorithm to classify a sample is 0, i.e., samples that exceed this threshold (i.e., score >0) are classified as likely to respond/could benefit from treatment comprising a KDM1A inhibitor.

In some embodiments, the classification algorithm is a Support Vector Machine (SVM). SVM is used for classification by mapping a training data set in space and constructing an N-dimensional hyperplane that optimally classifies sample data into two categories (e.g., sensitive and resistant to KDM1 Ai) for use as a threshold function. In addition to performing linear classification, SVMs may also perform nonlinear classification using a kernel technique to map sample data to a high-dimensional feature space. The new data are then mapped into the same space using the scoring function of the trained SVM and predicted to belong to a class according to which side of the hyperplane they fall. The performance of the classification algorithm (including the threshold) can be further evaluated by comparing the classification predicted by the algorithm to experimental values, calculating true positives, false positives, true negatives and false negatives, as well as sensitivity, specificity, etc., and optionally multiple rounds of training can be performed using training samples known to be responsive/resistant to KDM1Ai to adjust model parameters and/or optimize performance.

As used herein, the term "over" refers to: when the level or score of a test sample (as the case may be) is compared to a corresponding threshold, classifying the sample into a class of samples known to be sensitive to KDM1Ai, the biomarker level or score of the test sample, e.g., from a SCLC patient sample (unknown sensitivity to KDM1 Ai) under consideration for treatment with KDM1Ai, will exceed or cross the threshold. In some embodiments, the threshold is a cut-off value, and the biomarker level or score will exceed the threshold (cut-off value) when the biomarker level or score is higher than its corresponding cut-off value. Comparison of the level of a biomarker or score in a sample to a corresponding threshold value for the biomarker or score may be done mentally, manually, or automatically by a computer program.

As used herein, the term "sample" in relation to a patient sample for use in a method according to the invention may be a tumor sample (e.g. a biopsy sample, e.g. from a primary or metastatic SCLC lesion), a body fluid or a patient-derived cell line, a PDX sample ("PDX" refers to "patient-derived xenograft", i.e. a human tumor grown in mice) or one or more exosomes. Preferably, the sample is enriched/enriched (enriched) for tumor cells. A sample from a patient for carrying out a method according to the invention is obtained before the start of KDM1A inhibitor treatment (i.e. in the case of current no KDM1A inhibitor treatment and/or not, for example, after prior KDM1A inhibitor treatment (administration of KDM1A inhibitor) if the biomarker level is within a time period that is still adjustable by the (remaining) KDM1A inhibitor) -for example, a sample from a patient for use in a method according to the invention cannot be obtained within two weeks, preferably within one month, after prior KDM1A inhibitor treatment (or administration of KDM1A inhibitor). Biopsy samples can be obtained by well-known techniques and can be fresh, or can be processed by post-collection preparation and storage techniques (e.g., freezing, fixation, and/or embedding, such as formalin fixation, paraffin embedding, fresh flash freezing, fixation and frozen OCT embedding, etc.). Body fluid samples may be obtained by well-known techniques, including samples of blood, sputum, bronchoalveolar lavage fluid, or any other bodily secretion or derivative thereof that may contain SCLC cells. Isolated cells may be obtained from body fluids or tissues or organs by separation techniques such as centrifugation or cell sorting. Of course, the cell sample may be subjected to various well-known post-collection preparation and storage techniques (e.g., nucleic acid and/or protein extraction, fixation, storage, freezing, ultrafiltration, concentration, evaporation, centrifugation, etc.) prior to assessing the level of the marker in the sample. Preferably, the sample for measuring biomarker levels is enriched/enriched for the presence of SCLC cells or SCLC cell-derived vesicles (e.g. exosomes etc.). For example, SCLC cells can be isolated from sputum using methods described in the literature (Chest.1992, 8 months; 102(2): 372-4). SCLC Circulating Tumor Cells (CTCs) can be purified from blood by the methods described in the literature (Peeters et al, Br J Cancer 2013, 4.2; 108(6): 1358-67; Hodgkinson et al, Nat Med.2014 8; 20(8): 897-. Biomarker levels may also be analyzed from spatially defined regions of SCLC cell-enriched samples, which may be determined, for example, by an anatomical pathologist using standard methods used in the art.

In the context of treatment comprising a KDM1A inhibitor, the terms "responsive", "response", "sensitivity", "sensitive" and the like mean that the SCLC patient (or sample, SCLC cell line, etc.) shows a positive response to KDM1A inhibition, i.e. a positive response to treatment comprising a KDM1A inhibitor. In a more simplified form, the terms "responsive to treatment comprising a KDM1A inhibitor" and the like may be expressed as "responsive to a KDM1A inhibitor", "responsive to inhibition of KDM 1A", and the like. For example, "a positive response to a treatment comprising a KDM1A inhibitor" or "benefit from a treatment comprising a KDM1A inhibitor" may be or may include reversing, alleviating or inhibiting the progression of the disease SCLC or one or more symptoms thereof. As used herein, the term "more likely to respond" may refer to "more responsive" or simply "responsive".

The phrases "identifying a patient" or "selecting a patient" are used interchangeably and, as used herein, refer to the use of generated information or data relating to biomarker levels in a patient sample to identify or select patients more likely to respond to treatment comprising a KDM1A inhibitor (or benefit from treatment comprising a KDM1A inhibitor) or less likely to respond to treatment comprising a KDM1A inhibitor (or benefit from treatment comprising a KDM1A inhibitor). The information or data used or generated may be in any form, written, spoken, or electronic. The methods and uses provided herein may include communicating the results, information or data to the patient and/or any person involved in or responsible for the treatment of the patient, the treatment comprising a KDM1A inhibitor. In some embodiments, the use of the generated information or data includes communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof. In some embodiments, the communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof, is performed by a computing device, an analyzer unit, or a combination thereof. In some further embodiments, the communicating, presenting, reporting, storing, sending, transferring, supplying, transmitting, distributing, or a combination thereof is performed by a laboratory or medical professional. In some embodiments, the information or data comprises a comparison of biomarker levels to a threshold. In some embodiments, the information or data includes an indication that the patient is more likely or less likely to respond to (or benefit from) treatment comprising a KDM1A inhibitor.

As used herein, the phrase "predicting responsiveness of a patient" refers to the use of information or data generated relating to biomarker levels in a patient sample to assess the likelihood that the patient will respond to a treatment comprising an inhibitor of KDM 1A. The information or data used or generated may be in any form, written, spoken, or electronic. In some embodiments, the use of the generated information or data includes communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof. In some embodiments, the communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof, is performed by a computing device, an analyzer unit, or a combination thereof. In some further embodiments, the communicating, presenting, reporting, storing, sending, transferring, supplying, transmitting, distributing, or a combination thereof is performed by a laboratory or medical professional. In some embodiments, the information or data comprises a comparison of biomarker levels to a threshold. In some embodiments, the information or data includes an indication that the patient is more or less likely to respond to a treatment comprising an inhibitor of KDM 1A.

As used herein, the phrase "selecting a treatment" refers to using the generated information or data relating to biomarker levels in a patient sample to identify or select a treatment (therapy) for the patient. In some embodiments, the treatment may comprise an inhibitor of KDM 1A. The information or data used or generated may be in any form, written, spoken, or electronic. In some embodiments, the use of the generated information or data includes communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof. In some embodiments, the communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof, is performed by a computing device, an analyzer unit, or a combination thereof. In some further embodiments, the communicating, presenting, reporting, storing, sending, transferring, supplying, transmitting, distributing, or a combination thereof is performed by a laboratory or medical professional. In some embodiments, the information or data comprises a comparison of biomarker levels to a threshold. In some embodiments, the information or data includes an indication that a treatment comprising a KDM1A inhibitor is appropriate for the patient (i.e., the patient is likely to respond to the treatment).

As used herein, the phrase "recommending a treatment" refers to using the generated information or data relating to biomarker levels to recommend or select a treatment comprising a KDM1A inhibitor for a patient identified or selected as more likely or less likely to respond to a treatment comprising a KDM1A inhibitor. The information or data used or generated may be in any form, written, spoken, or electronic. In some embodiments, the use of the generated information or data includes communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof. In some embodiments, the communicating, presenting, reporting, storing, sending, transferring, provisioning, transmitting, allocating, or a combination thereof, is performed by a computing device, an analyzer unit, or a combination thereof. In some further embodiments, the communicating, presenting, reporting, storing, sending, transferring, supplying, transmitting, distributing, or a combination thereof is performed by a laboratory or medical professional. In some embodiments, the information or data comprises a comparison of biomarker levels to a threshold. In some embodiments, the information or data includes an indication that a treatment comprising a KDM1A inhibitor is appropriate for the patient.

As used herein, a "kit" is any article of manufacture (e.g., a package or container) containing one or more reagents for measuring ASCL1 and SOX2 levels as described herein that are promoted, distributed, or sold as a unit for performing the methods of the invention.

As used herein, "reagents" for measuring levels of ASCL1 or SOX2 include any reagents commonly used in the art for measuring levels of biomarkers, including, but not limited to, antibodies specifically recognizing ASCL1 or SOX2 proteins, or probes and/or primers that hybridize to ASCL1 or SOX2 polynucleotides for specifically detecting biomarkers according to the invention, and any other such reagents associated with the methods for measuring levels of biomarkers described in more detail elsewhere herein.

The present invention provides tools and methods for identifying SCLC patients who have an increased likelihood of responding to treatment with a KDM1A inhibitor and are therefore most suitable for treatment with a KDM1A inhibitor, and methods of treatment of these patients using a KDM1A inhibitor. The present invention is based, at least in part, on the following findings: the levels of ASCL1 and SOX2 can be used as biomarkers (e.g., predictive biomarkers) in methods of predicting the likelihood of response to KDM1A inhibitor therapy. As described herein and in the accompanying examples, the inventors have found that high levels of ASCL1 and SOX2 in SCLC cell lines correlate with the responsiveness (sensitivity) of SCLC to treatment with KDM1A inhibitors. As shown in examples 2, 3 and 4 and fig. 1-3, SCLC cell lines expressing high levels of ASCL1 and SOX2 are generally responsive (sensitive) to KDM1A inhibitors, while SCLC cell lines exhibiting low levels of one or both of ASCL1 and SOX2 are generally resistant to KDM1A inhibition therapy. Thus, ASCL1 and SOX2 levels can be used to stratify patients with SCLC treated with KDM1A inhibitor, identifying those more likely to respond to KDM1A inhibitor treatment and those less likely to respond to KDM1A inhibition. The method according to the invention using ASLC1 and SOX2 enables prediction of SCLC responsiveness to KDM1A inhibition with high sensitivity and specificity, which is very interesting in view of the reduced number of biomarkers used, as shown in more detail in the appended examples. The method according to the invention can measure biomarkers as mRNA levels or protein levels since a good correlation between their mRNA and protein expression levels is shown as in example 5 using SCLC cell lines or patient derived samples such as SCLC PDX samples. This makes the method according to the invention particularly advantageous for use in clinical practice, especially in hospitals, since the most easily obtained samples from cancer patients are usually fixed tumor biopsies, which are more suitable for protein level analysis, since standard biopsy sample fixation procedures are known to fragment and reduce RNA.

As previously mentioned, the method of the present invention includes measuring the levels of ASCL1 and SOX 2. These markers are known per se in the art and are also described below.

Public database directories of ASCL1 and SOX 2:

the DNA and protein sequences of human ASCL1 and SOX2 have been previously reported, see GenBank accession numbers listed below (NCBI-GenBank Flat File Release 225.0,2018, month 4, 15) and UniProtKB/Swiss-Prot accession numbers (Knowledgebase Release 2018 — 04,2018, month 4, 25), each of which is incorporated herein by reference in its entirety for all purposes. Such sequences can be used to design steps for measuring ASCL1 and SOX2 levels by methods known to those skilled in the art.

Exemplary nucleotide and amino acid sequences of human ASCL1 and SOX2 are shown herein in SEQ ID NOs 1 through 4. The following table assigns markers and corresponding sequences:

in one aspect, the invention provides a method of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold. Thus, in some embodiments, the invention provides a method of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor, wherein the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the invention provides a method of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from a patient prior to initiating treatment comprising a KDM1A inhibitor, and using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein a patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In another aspect, the invention provides a method of identifying SCLC patients who may benefit from treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as a patient that may benefit from treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein when the score in the sample exceeds a threshold, the patient is identified as a patient that may benefit from treatment comprising a KDM1A inhibitor. Thus, in some embodiments, the invention features a method of identifying an SCLC patient who may benefit from treatment comprising a KDM1A inhibitor, the method comprising measuring levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor, wherein the patient is identified as benefiting from treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the invention features a method of identifying an SCLC patient who may benefit from treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor, and using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as benefiting from treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the invention provides a method of predicting the responsiveness of an SCLC patient to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to the commencement of the treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to be responsive to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold. Accordingly, in some embodiments, the present invention provides a method of predicting responsiveness of an SCLC patient to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating the treatment comprising the KDM1A inhibitor, wherein the patient is identified as more likely to be responsive to the treatment comprising the KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the invention provides a method of predicting responsiveness of an SCLC patient to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor, and using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to be responsive to treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the present invention provides a method of assessing the likelihood of a SCLC patient responding to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold. Accordingly, in some embodiments, the present invention provides a method of assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor, wherein the patient is identified as more likely to respond to the treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the present invention provides a method of assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment comprising a KDM1A inhibitor, and using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the invention provides a method of assessing the likelihood of a patient's SCLC responding to a treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from a SCLC patient prior to the commencement of treatment comprising a KDM1A inhibitor. In some embodiments, SCLC is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein SCLC is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold. Accordingly, in some embodiments, the present invention provides a method of assessing the likelihood of SCLC responding to treatment comprising a KDM1A inhibitor, the method comprising measuring ASCL1 and SOX2 levels in a sample from a SCLC patient prior to initiating treatment comprising a KDM1A inhibitor, wherein SCLC is identified as more likely to respond to treatment comprising a KDM1A inhibitor when the levels of each of ASCL1 and SOX2 in the sample exceed a threshold. In some embodiments, the present invention provides a method of assessing the likelihood of SCLC responding to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from an SCLC patient prior to initiating treatment comprising a KDM1A inhibitor, and using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein SCLC is identified as more likely to respond to treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

In a further aspect, the invention provides a method of selecting a treatment for an SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating the treatment. In some embodiments, when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold, the method comprises providing a recommendation that the selected treatment for the patient comprise a KDM1A inhibitor. In some embodiments, the methods further comprise using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, and providing a recommendation that the selected treatment for the patient comprise a KDM1A inhibitor when the score in the sample exceeds a threshold. Thus, in some embodiments, the invention also provides a method of selecting a treatment for a SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating the treatment, and providing a recommendation that the selected treatment for the patient comprises a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold. In some embodiments, the invention also provides a method of selecting a treatment for a SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating treatment, using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, and providing a recommendation that the selected treatment for the patient comprise a KDM1A inhibitor when the score in the sample exceeds a threshold.

All the above methods according to the invention comprise measuring the level of the biomarker of the invention (ASCL1 and SOX2) in the sample and assessing the biomarker level or derived score (based on said level) vs. threshold. Typically, each biomarker (i.e., ASCL1 and SOX2) has its threshold (which may be established as described elsewhere herein), and the (measured) levels of ASCL1 and SOX2 are each evaluated relative to their respective thresholds, wherein when the level of each of ASCL1 and SOX2 in the sample exceeds its threshold, the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor (or, if applicable, as a patient who may benefit from treatment comprising a KDM1A inhibitor or the like). Alternatively, the (measured) levels of ASCL1 and SOX2 in a sample can be used using a classification algorithm to generate a score for the sample; in such cases, a score threshold (which may be established as described elsewhere herein) will be applied, and when the score in the sample exceeds the threshold, the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor (or, if applicable, as a patient who may benefit from treatment comprising a KDM1A inhibitor or the like).

In some embodiments of any of the preceding aspects, the method further comprises the step of obtaining or providing a sample from the patient. The obtaining/providing step precedes the measurement of the level of the biomarker (and prior to administering any treatment comprising a KDM1A inhibitor to the patient from whom the sample is to be obtained/provided).

In some embodiments of any of the preceding aspects, if the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor, the method further comprises recommending, prescribing or administering to the patient a therapeutically effective amount of a treatment comprising a KDM1A inhibitor.

In some embodiments of any of the preceding aspects, if the patient is identified as unlikely to respond to a treatment comprising a KDM1A inhibitor, the method may optionally further comprise recommending that the patient not be treated with a KDM1A inhibitor.

In the methods according to the invention, the level of ASCL1 and SOX2 can be determined at the mRNA level or protein level using any method known in the art for measuring mRNA or protein levels, including methods as described herein.

In the method according to the invention, mRNA from the sample can be used directly for determining the level of the biomarker. In the method according to the invention, said level may be determined by hybridization. In the method according to the invention, RNA may be converted into cDNA (complementary DNA) copies using methods known in the art. Detection methods may include, but are not limited to, quantitative reverse transcriptase polymerase chain reaction (qRT-PCR), gene expression analysis, RNA sequencing, nanopore sequencing, microarray analysis, gene expression chip analysis, (in situ) hybridization techniques, RNAscope and chromatography, as well as any other technique known in The art, such as those described in Ralph multiplex, "The Nucleic Acid Protocols Handbook", published 2000, ISBN: 978-0-89603-. Methods for detecting RNA can include, but are not limited to, PCR, real-time PCR, digital PCR, hybridization, microarray analysis, and any other technique known in the art, such as those described in Leland et al, "Handbook of Molecular and cellular Methods in Biology and Medicine", published 2011, ISBN 9781420069389.

In the method according to the invention, the method may comprise detecting the protein expression level of the biomarker. Any suitable method of Protein detection, quantification and comparison may be used, such as those described in John m.walker, "The Protein Protocols Handbook", published in 2009, ISBN 978-1-59745-198-7. Protein expression levels of biomarkers can be determined by immunoassays comprising recognition of a protein or protein complex by an antibody or antibody fragment, including but not limited to enzyme-linked immunosorbent assay (ELISA), "sandwich" immunoassays, immunoradiometric assays, in situ immunoassays, alphaLISA immunoassays, protein proximity assays, proximity ligation assay techniques (e.g., protein qPCR), western blot analysis, immunoprecipitation assays, immunofluorescence assays, flow cytometry, immunization, and protein binding assaysHistochemistry (IHC), immunoelectrophoresis, protein immunostaining, confocal microscopy; or by similar methods, wherein the antibody or antibody fragment is replaced by a chemical probe, aptamer, receptor, interacting protein, or any other biomolecule that recognizes a biomarker protein in a specific manner; or by

Fluorescence Resonance Energy Transfer (FRET), Differential Scanning Fluorescence (DSF), microfluidics, spectrophotometry, mass spectrometry, enzymatic assays, surface plasmon resonance, or a combination thereof. The immunoassay may be a homogeneous assay or a heterogeneous assay. In homogeneous assays, the immune response typically includes specific antibodies, labeled analyte, and sample of interest. After the antibody binds to the labeled analyte, the signal generated by the label is modified, either directly or indirectly. The immune response and its degree of detection can be performed in homogeneous solution. Immunochemical labels which may be used include free radicals, radioisotopes, fluorescent dyes, enzymes, bacteriophages or coenzymes. In heterogeneous assays, reagents are typically a sample, an antibody, and a means to produce a detectable signal. The antibody may be immobilized on a support, such as a bead, plate or slide, and contacted with a sample suspected of containing the antigen in a liquid phase. The support is then separated from the liquid phase and the support phase or liquid phase is examined for such detectable signal using a signal generating means. The signal is related to the presence of the analyte in the sample. Means for generating a detectable signal include the use of radioactive, fluorescent, or enzymatic labels.

In the method according to the invention, antibodies directed against the biomarker of interest may be used. In the method according to the present invention, a kit for detection may be used. Such antibodies and kits are commercially available, e.g., EMD Millipore, R for biochemical assays&D Systems, Thermo Scientific Pierce Antibodies, Novus Biologicals, Aviva Systems Biology, Abnova Corporation, AbD Serotec, or others. Alternatively, the antibody may be synthesized by any known method. As used herein, the term "antibody" is intended to include monoclonal antibodies, polyclonal antibodies, or fragments thereof,Single chain antibodies and chimeric antibodies. The antibodies can be conjugated to a suitable solid support (e.g., beads such as protein a or protein G agarose, microspheres, plates, slides, or wells formed from materials such as latex or polystyrene) according to known techniques, e.g., passive binding. An antibody as described herein may likewise be conjugated to a detectable label or capable of generating a signal such as a radiolabel (e.g.,³⁵s), enzyme labels (e.g., horseradish peroxidase, alkaline phosphatase), fluorescent labels (e.g., fluorescein, Alexa, green fluorescent protein, rhodamine), phthalocyanine-containing beads that release singlet oxygen upon illumination at 680nM and emit excitation light after its subsequent absorption by europium-or erbium-containing acceptor beads, and oligonucleotide-labeled moieties conjugated. The label may generate a signal directly or indirectly. The generated signal may comprise, for example, fluorescence, radioactivity or luminescence, according to known techniques.

Alternative protein capture agents including aptamers, affimers or chemical probes (chemoprobes) with high affinity and selectivity for the protein biomarker to be analyzed may be substituted for the antibody.

In some embodiments, such as in immunofluorescence analysis of biopsies, when measuring the protein level of a biomarker, the level of the biomarker can be assessed in a portion of the SCLC tumor cells, such as in the nucleus of the tumor cells.

The level of a biomarker may be expressed in any form of mRNA expression measurement or protein expression measurement used in the art, and may be raw data or processed data, i.e., the raw data is background subtracted, normalized, or otherwise converted using corrections or other mathematical operations commonly used in the art. For example, when measured by microarray hybridization, biomarker levels can be expressed as the hybridization signal intensity value of the sample, Log2 (the hybridization intensity value of the sample), or Log2 (the hybridization signal intensity value of the sample/the hybridization signal intensity value of the reference sample). Examples of suitable reference samples are patient-derived tumor samples with high expression levels of ASCL1 and SOX2 obtained from xenografts or PDX models or SCLC cell pellets. Can be obtained by using a single color or 2-color hybridizationHybridization signal values. For example, when measured by qRT-PCR, biomarker levels can be expressed as cross-point-PCR-cycle (Cp) values (expressed as the number of cycles required to achieve a particular detection threshold level of amplification-related fluorescence), Δ Cp ═ C_p–C_{p, reference gene}、2^-CpValue or 2^-ΔCpThe value is obtained. For example, biomarker levels can be expressed as read length Per Million (Reads Per Million, RPM), read length Per Kilobase Per Million (RPKM), fragment Per Kilobase Per Million (FPKM), or Transcript Per Million (TPM) values, as measured by RNA sequencing. For example, when measured by western blot, biomarker levels may be expressed as integrated density (A.U) of the corresponding bands after image analysis, either as raw integrated density or normalized to protein content and/or as a ratio relative to a reference sample. For example, when measured by immunostaining, biomarker levels can be expressed as the integrated density of nuclear signals (A.U) after image analysis, as the raw integrated density (A.U)/area unit (pixel)²Or μm²) Either as an integrated density/kernel or as a ratio with respect to a reference sample. For example, when measured by ELISA, biomarker levels can be expressed as r.l.u (relative light units) or absorbance units as raw (values) or background corrected (values) or normalized to total protein content and/or as a ratio relative to a reference sample.

In some embodiments of any of the methods according to the invention, the biomarker levels (i.e., ASCL1 levels and SOX2 levels) are mRNA expression levels. Preferably, mRNA expression levels are measured by qRT-PCR.

In some embodiments of any of the methods according to the invention, the biomarker level is a protein expression level. Preferably, the protein expression level is measured by fluorescence immunohistochemistry.

In some embodiments, the level of biomarker expression in the immunofluorescent stain is visually classified as high, medium, low, or undetectable based on the level of staining intensity, with values of 3, 2, 1, and 0, respectively. Samples at medium and high levels (values 2 and 3) are considered "positive" (i.e. above the threshold for the respective biomarker), while samples at undetectable or low levels (values 0 and 1) are considered "negative" (i.e. not above the threshold for the respective biomarker). When the sample is considered "positive" for both biomarkers (i.e., when the levels of ASCL1 and SOX2 in the sample are each classified as levels 2 or 3), the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor/the patient is identified as he/she may benefit from treatment comprising a KDM1A inhibitor.

Alternatively, in some embodiments, the expression level of a biomarker in an immunofluorescent stained image may be quantified. DNA dyes such as DAPI staining can be used to localize nuclei quantitatively using fluorescence. Quantitative analysis of SOX2 and ASCL1 expression in the nucleus of cells can be performed using immunofluorescence. Individual images from the biomarkers and DAPI staining can be analyzed using imaging software, for example using ImageJ. The signal can be obtained by background subtraction and normalization against a reference (calibration) sample. Suitable calibration samples are samples with high and uniform nuclear expression of both biomarkers, for example a xenograft sample of NCI-H1417 origin. The normalized quantitative value may be expressed as%, or as a ratio relative to a calibration sample. The threshold for each biomarker can be established as the fraction of the signal of the calibration sample (fraction) and should be chosen higher than the (average) signal of the negative control sample used (which may be, for example, a normal lung biopsy sample or a xenograft sample in which both biomarkers are under-expressed or undetectable). Preferably, the threshold is at least the mean signal of the negative control sample plus 1SD, 2SD or 3SD, where SD means the standard deviation.

In some embodiments of the methods and uses according to the invention, the term "when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold value" or the like may mean "when the level of each of ASCL1 and SOX2 in the sample is increased as compared to a control". In this case, when the level of each of ASCL1 and SOX2 in the sample is increased compared to the control, the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor/the patient is identified as he/she may benefit from treatment comprising a KDM1A inhibitor. The terms "control" or "reference" are used interchangeably herein. Non-limiting examples of "control" (e.g., "control value") or "reference" (e.g., "reference value") can be ASCL1 and SOX2 levels in a sample or pool of samples from one or more healthy individuals/subjects, respectively. For example, a healthy individual/subject may be an individual/subject not having SCLC as defined herein, in particular an individual/subject not having SCLC at the time the sample is obtained from the individual/subject. Alternatively, for example, a healthy individual/subject may be an individual/subject that does not have a disease or disorder associated with elevated levels of each of ASCL1 and SOX 2. Preferably, the healthy individual/subject is a human. Another non-limiting example of a "control" (e.g., "control value") or "reference" (e.g., "reference value") may be ASCL1 and SOX2 levels, respectively, in a sample or sample pool from a "non-responder," e.g., from one or more patients who have SCLC and are known to be non-responsive to a KDM1A inhibitor. Another example of a "non-responder" control is a cell line/cell/tissue that shows no response to a KDM1A inhibitor in an in vitro, ex vivo, or (patient-derived) xenograft test. Another non-limiting example of a "control" is an "internal standard", e.g. a purified or synthetically produced protein and/or peptide or mixture thereof, or corresponding nucleic acid, wherein the amount of each protein/peptide (or corresponding nucleic acid) is calibrated by using the above-mentioned "non-responder" control. In particular, this "internal standard" may comprise the proteins ASCL1 and SOX2 (or corresponding nucleic acids) as described and defined herein. Non-limiting examples of "control" (e.g., "control value") or "reference" (e.g., "reference value") may be the level of ASCL1 and SOX2, respectively, from a patient sample to be identified herein, if, for example, the sample is obtained before the patient has SCLC, before the patient is predisposed (or at risk) of having SCLC cancer, or if the sample is obtained when the patient (completely) recovers from a previous SCLC.

In some embodiments of any of the preceding aspects, the sample is an SCLC biopsy sample, preferably an SCLC biopsy sample enriched for SCLC cells.

Preferably, in any of the methods of the invention, the patient is a human patient.

Preferably herein, the above (diagnostic) method is an in vitro method. As used herein, "in vitro" refers to methods of the invention as described above, e.g., methods of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, etc., not performed in vivo, i.e., not directly on the patient, but in vitro in a living human (or other mammal), on samples obtained from the patient and isolated/isolated from the patient (i.e., taken from its in vivo location).

By using the above methods, a subset of SCLC patients with a higher chance of responding, i.e., more likely to respond to receiving treatment comprising KDM1Ai or benefit from receiving treatment comprising KDM1Ai, can be identified. The invention also relates to therapeutic methods for treating those SCLC patients who have undergone such identification.

Thus, in a further aspect, the invention provides a method of treating a SCLC patient comprising administering to the patient a therapeutically effective amount of a treatment comprising a KDM1A inhibitor if the patient has been identified as more likely to respond to treatment comprising a KDM1A inhibitor using a method according to any preceding aspect before commencing treatment comprising a KDM1A inhibitor.

In some embodiments of any of the preceding aspects, the method further comprises the step of obtaining or providing a sample from the patient.

Preferably, in any of the methods of treatment and uses according to the present invention, the patient is a human patient.

The same applies elsewhere in this disclosure relating to methods of measuring biomarker levels, sample types, etc. to the above-described therapeutic methods and related therapeutic uses.

KDM1A inhibitors that may be used according to the invention include any KDM1A inhibitor currently known in the art or that may be reported in the future. Preferably, KDM1A inhibitors are small molecules. Irreversible and reversible KDM1A inhibitors have been reported. Irreversible KDM1A inhibitors exert their inhibitory activity by covalently binding to the FAD cofactor within the active site of KDM1A and are typically based on 2-cyclyl-cyclopropylamino moieties, such as 2- (hetero) aryl-cyclopropylamino moieties. Reversible KDM1A inhibitors have also been disclosed. Preferably, KDM1A inhibitors should be active in cells. For example, the cellular activity of a KDM1A inhibitor may be determined using an established in vitro cellular assay for the KDM1A inhibitor, such as, for example, an SCLC cell viability assay (e.g., an assay as described in example 1 herein or Mohammad et al, 2015, supra) or an acute myeloid leukemia cell line differentiation assay (e.g., an assay as described in Lynch et al, Anal biochem.2013, 11 months 1; 442(1):104-6.doi: 10.1016/j.ab.2013.07.032).

Examples of KDM1A inhibitors that may be used according to the invention include, but are not limited to, those disclosed below: WO2010/, WO2011/, WO2012/, WO2013/, WO2010/, WO2011/, WO2012/, WO2013/, WO2014/, WO2015/, WO2007/, WO2008/, WO 2014/WO 2015/089192, WO2015/, WO2016/130952, WO2016/177656, WO2017/, WO 2012/469, WO 2012013/, WO2014/, WO2016, WO2015/134973, WO2015/168466, WO2015/200843, WO2016/003917, WO2016/004105, WO2016/007722, WO2016/007727, WO2016/007731, WO2016/007736, WO2016/034946, WO2016/037005, WO2016/161282, WO2016172496WO2017/004519, WO2017/027678, WO2017/079476, WO2017/079670, WO2017/090756, WO2017/109061, WO 2010337/116558, WO2017/114497, WO2017/149463, WO2017/157322, WO2017/195216, WO2017/198780, WO2017/215464, WO2018/081342, WO2018/081343, US-0324147, US2015-0065434, US 2017-148 0283397, CN 106862, CN 104280, CN 1041340, CN 1031071340, CN 103311340, CN 103311073163, CN 10331107311079, CN 923110747779, CN 1069810747, CN 106989, CN 1069810747779, CN 10698107106989, CN 1069810747, CN 9810747, CN 989, CN 9810747, CN 989, CN 9810747, CN 3, CN 9810747, CN 1069810747, CN 3, CN 1064738, CN 983, CN 9810747, CN 3, CN 98311073, CN 1064738, CN 3, CN 1064738, CN 1064710847, CN 3, CN 1064738, CN 3, CN 10647779, CN 1064710847, CN 1064738, CN 3, CN 1064768, CN 3, CN 1064738, CN 1064768, CN 3, CN 9867, CN 3, CN 1064738, CN

Or

Including any optically active stereoisomer thereof or any pharmaceutically acceptable salt thereof.

A particularly preferred KDM1A inhibitor is (trans) -N1- ((1R,2S) -2-phenylcyclopropyl) cyclohexane-1, 4-diamine [ CAS registry No. 1431304-21-0] or a pharmaceutically acceptable salt thereof, more preferably (trans) -N1- ((1R,2S) -2-phenylcyclopropyl) cyclohexane-1, 4-diamine dihydrochloride [ CAS registry No. 1431303-72-8 ]. The compound (trans) -N1- ((1R,2S) -2-phenylcyclopropyl) cyclohexane-1, 4-diamine, which in turn may be named (1R,4S) -N1- ((1R,2S) -2-phenylcyclopropyl) cyclohexane-1, 4-diamine, is known as ary-1001 or iadademstat, has the chemical structure shown below:

ORY-1001 is disclosed, for example, in WO2013/057322, see example 5 therein. Pharmaceutical formulations comprising ORY-1001 for administration to a patient may be prepared according to methods known to those skilled in the art, for example as described in WO 2013/057322.

KDM1A inhibitors may be administered as the API alone, i.e., as a monotherapy, or may be administered in combination with one or more other APIs, e.g., other anti-cancer agents used to treat SCLC.

While it is possible to administer KDM1A inhibitors (or treatments comprising KDM1A inhibitors) directly for therapy, typically KDM1A inhibitors are administered in the form of a pharmaceutical composition comprising a compound as an active pharmaceutical ingredient together with one or more pharmaceutically acceptable excipients or carriers. Any reference herein to an inhibitor of KDM1A includes reference to the compound per se, i.e. the corresponding compound in its non-salt form (e.g. as the free base) or in any pharmaceutically acceptable salt or solvate form thereof, as well as to pharmaceutical compositions comprising said compound (or a pharmaceutically acceptable salt or solvate thereof) and one or more pharmaceutically acceptable excipients or carriers.

KDM1A inhibitors may be administered by any means that achieves the intended purpose. Examples include administration by oral or parenteral (including, for example, intravenous or subcutaneous) routes.

For oral delivery, the compounds may be incorporated into formulations comprising pharmaceutically acceptable carriers such as binders (e.g., gelatin, cellulose, tragacanth), excipients (e.g., starch, lactose), lubricants (e.g., magnesium stearate, silicon dioxide), disintegrants (e.g., alginates, Primogel and corn starch), and sweetening or flavoring agents (e.g., glucose, sucrose, saccharin, methyl salicylate and peppermint). The formulation may be delivered orally, for example in the form of a closed gelatin capsule or a compressed tablet. Capsules and tablets may be prepared by any conventional technique. Capsules and tablets may also be coated with various coatings known in the art to improve the flavor, taste, color, and shape of the capsules and tablets. In addition, a liquid carrier such as a fatty oil may also be included in the capsule.

Suitable oral formulations may also be in the form of suspensions, syrups, chewing gums, wafers, elixirs and the like. Conventional agents for improving the flavor, taste, color and shape of particular forms may also be included if desired. In addition, for convenience of administration via enteral feeding tubes to patients who cannot swallow, the active compounds may be dissolved in acceptable lipophilic vegetable oil carriers, such as olive oil, corn oil and safflower oil.

The compounds may also be administered parenterally in the form of solutions or suspensions, or may be in lyophilized form which can be converted to a solution or suspension form prior to use. In such formulations, diluents or pharmaceutically acceptable carriers may be used, such as sterile water and physiological saline buffer. Other conventional solvents, pH buffers, stabilizers, antimicrobials, surfactants, and antioxidants may be included. For example, useful components include sodium chloride, acetate, citrate or phosphate buffers, glycerol, glucose, fixed oils, methylparaben, polyethylene glycol, propylene glycol, sodium bisulfate, benzyl alcohol, ascorbic acid, and the like. The parenteral formulation may be stored in any conventional container, such as vials and ampoules.

Pharmaceutical compositions, such as oral and parenteral compositions, can be formulated in unit dosage form for ease of administration and uniformity of dosage. As used herein, "unit dosage form" refers to physically discrete units suitable as unitary dosages for subjects, each unit containing a predetermined quantity of active ingredient calculated to produce the desired therapeutic effect, in association with one or more suitable pharmaceutical carriers.

In therapeutic applications, the pharmaceutical composition will be administered in a manner appropriate to the disease to be treated, as determined by one of skill in the medical arts. Suitable dosages and suitable durations and frequencies of administration will depend on the following factors: such as the status of the patient, the severity of the disease, the particular KDM1A inhibitor administered, the method of administration, and the judgment of the attending physician. Generally, suitable dosages and administration regimens provide the pharmaceutical composition in an amount sufficient to provide a therapeutic benefit, e.g., improved clinical outcome, such as more frequent complete or partial remission, or longer disease free and/or overall survival, or reduction in severity of symptoms, or any other objectively identifiable improvement as indicated by the clinician. Effective dosages can generally be assessed or extrapolated using experimental models, such as dose response curves derived from in vitro or animal model test systems. One skilled in the art would be able to determine the appropriate dosage and treatment regimen based on the factors described above.

The pharmaceutical compositions of the present invention may be contained in a container, package or dispenser together with instructions for administration.

Examples

The following examples are provided to illustrate the invention. They should not be considered as limiting the scope of the invention, but merely as being representative thereof.

Example 1: classification of KDM1A inhibitor sensitive and resistant cell lines

For biomarker identification, KDM1A inhibitor ori-1001 (as described elsewhere herein) was classified for its response to KDM1A inhibitor treatment against seven SCLC cell lines based on viability assay results performed 4 or 10 days after treatment. For the 4-day viability assay, the supplier-recommended optimized medium supplemented with ORY-1001 for each cell line seeded with SCLC cell lines in 384-well plates at a final volume of 40. mu.L (maximum concentration: 50. mu.M; tested18 series of 1:2 dilutions). At 37 deg.C, 5% CO₂After 4 days incubation in controlled atmosphere, use

Assays cell viability was assessed according to the manufacturer's protocol (Promega). EC50 values were calculated using Microsoft Excel software for untreated cells (100% growth) and none

The reagents (100% growth inhibition) were normalized. After exposure to ORY-100110 days, viability quantification was performed in a 96-well plate format (maximum concentration 1. mu.M, NCI-H187 cell line 100 nM). Cells were initially seeded in 100. mu.L of medium (RPMI-164010% FBS 2mM glutamine, prepared by adding 4.5mM glutamine, 0.005mg/mL insulin, 0.01mg/mL transferrin, 30nM sodium selenite, 10nM hydrocortisone, 10nM beta-estradiol to DMEM: F125% FBS, except that NCI-H1876 cell line was cultured in HITES medium) and at 37 ℃ and 5% CO₂-incubation in a controlled atmosphere. On day 6, an additional 100. mu.L of medium containing ORY-1001 was added. After another 4 days, Alamar was used

The assay (Thermo Fisher Scientific) measures residual viability. After background subtraction, normalization was performed on vector-treated cells. EC50 values were calculated after fitting the nonlinear model using GraphPad Prism software.

Cell lines were classified as sensitive to KDM1A inhibition when growth inhibition after treatment with ary-1001 was equal to or greater than 25% and EC50 was less than 10 nM. SHP77 was classified as partially sensitive because although it was not sensitive under the test conditions herein, it was reported to be sensitive to ary-1001 in other assays. The results obtained are shown in table 1 below. The concentration at which 50% growth reduction was achieved (EC50) and the maximum percent growth inhibition (maximum% response) at the highest dose tested are reported in the table.

Table 1:

example 2 identification of ASCL1 and SOX 2as biomarkers of responsiveness to KDM1A inhibitors

To identify genes that can distinguish between SCLC cell lines sensitive and resistant to KDM1A inhibitor treatment, RNASeq analysis was performed on four SCLC cell lines identified as KDM1Ai sensitive, one identified as KDM1Ai partially sensitive and two identified as KDM1Ai resistant. Detailed information regarding their responsiveness to KDM1Ai and classification is provided in example 1 above.

Cell pellet for gene expression analysis

Cells were grown in flasks for 6 consecutive days. On the last day of the assay, cells were collected in Falcon tubes, counted, and then centrifuged at 1200rpm for 4 minutes. The supernatant was discarded, and the cells were suspended in 1mL PBS and transferred to an eppendorf sample and centrifuged at 3000rpm for 5 minutes in an eppendorf centrifuge at 4 ℃. Finally, the supernatant was discarded and the pellet was frozen at-80 ℃.

RNA sequencing

Total RNA was isolated from the samples using the QIAGEN miRNeasy Mini kit. RNA isolation was performed using QIAgen RNeasy Mini kit. The RNeasy Mini kit combines the selective binding properties of silica gel based membranes with the speed of micro spin technology. Briefly, tissues were homogenized and lysed in RLT buffer (containing β -mercaptoethanol) using a Lysing Matrix D tube (MP Biomedicals LLC) or vortexed. Ethanol was added to the homogenate to provide suitable binding conditions for all RNA molecules longer than 200 nucleotides (nt). Samples were loaded onto RNeasy Mini columns where total RNA bound to the membrane and contaminants were effectively washed away. High quality RNA was then eluted with nuclease-free water. The generated RNA was quantified using an Agilent Bioanalyzer and the integrity was evaluated.

Library generation was done using Illumina TruSeq Stranded mRNA Library Preparation. Cluster generation and sequencing of the library was performed on an Illumina HiSeq. Total RNA samples were converted to cDNA libraries using the TruSeq Stranded mRNA Sample Prep kit (Illumina). Starting from 100ng total RNA, polyadenylated RNA (mainly mRNA) was selected and purified using oligo-dT conjugated magnetic beads. This mRNA was chemically fragmented and converted to single-stranded cDNA using reverse transcriptase and random hexamer primers, and actinomycin D was added to inhibit DNA-dependent synthesis of the second strand. Double-stranded cDNA is generated by removing the RNA template and synthesizing the second strand in the presence of dUTP instead of dTTP. A single a base is added to the 3' end to facilitate ligation of sequencing adaptors, including a single T base overhang. The adaptor-ligated cDNA is amplified by polymerase chain reaction to increase the amount of sequence-ready library. During this amplification, the polymerase stops when it encounters a U base, making the second strand a poor template. Thus, the amplification material uses the first strand as a template, thereby retaining information about that strand. The size distribution of the final cDNA library was analyzed using an Agilent Bioanalyzer (DNA 1000 kit), quantified by qPCR (KAPA library quantification kit), and then normalized to 2nM to prepare for sequencing. Standard Cluster Generation kit v5(Standard Cluster Generation kit v5) binds cDNA libraries to the surface of the flow cell. cBot amplifies the ligated cDNA constructs isothermally so that each yields a colony of about 1000 copies. The DNA sequence was determined directly by TruSeq SBS kit using sequencing-by-synthesis techniques.

The following quality control indicators were applied: the sample had 100ng of input RNA and RIN values ≧ 7.0 to facilitate library preparation. At least 3000 million paired-end reads of 50bp in total were generated per single sample. After subtracting out various off-target sequences such as ribosomal RNA, phiX, homopolymer repeats and globin RNA, at least 2850 ten thousand read lengths are provided.

Illumina HiSeq software reports the total number of clusters (DNA fragments) loaded in each lane, the percentage passed through the sequencing quality filter (identifying errors due to overload and sequencing chemistry), the phred quality score of each base for each sequence read, the overall average phred score for each sequencing cycle, and the overall error percentage (based on alignment to the reference genome). For each RNA-seq sample, the percentage of reads containing mitochondrial and ribosomal RNAs was calculated. The FASTQC package was used to provide additional QC-metrics (base distribution, sequence repeats, over-represented sequences and enriched kmer) and graphical summaries. The original reads were aligned to the human genome (hg19) using GSNAP and recommended RNASeq data options. In addition to genomic sequences, GSNAP also provides a database of human splice junctions and transcripts based on Ensembl v 73. The generated SAM file is then converted into a classified BAM file using Samtools. Gene expression values were calculated as RPKM values and read length counts according to Mortazavi et al (Nat Methods (2008)5(7): 621-8). Normalized read length counts were obtained using the R-packet DESeq 2. Data are reported as the average of Log2(RPKM) from three independent experiments. RPKM represents reads per million per kilobase.

Results

Gene expression of ASCL1 and SOX2 in SCLC cell lines measured by RNASeq are shown in table 2 below. The table shows the mean (Av.) and Standard Deviation (SD) in Log2 (RPKM). The color code displayed is based on gene expression level; the darker the color, the higher the expression of the biomarker. For comparison, GAPDH expression was reported in parallel to show stable expression of the reference gene across samples.

Table 2:

these results are graphically represented in fig. 1, which is a dot plot showing the expression of ASCL1 (Y-axis) and SOX2 (X-axis) of KDM1A inhibition sensitive, partially sensitive or resistant SCLC cell lines as described above, measured by RNA-seq.

Based on RNASeq data generated for these cell lines, it was determined that all ary-1001 sensitive and partially sensitive cell lines expressed high levels of ASCL1 and moderate to high levels of SOX2, whereas in SCLC cell lines that were resistant to ary-1001 treatment, very low levels of ASCL1 or SOX2 were detected (Log2(RPKM) ≦ 0) as shown in table 2 and fig. 1. Thus, ASCL1 and SOX2 may be used as biomarkers to identify SCLC cells or subjects with SCLC that are sensitive (i.e., responsive) or more likely to be sensitive (responsive) to treatment with KDM1A inhibitors (e.g., ary-1001).

Example 3 validation of ASCL1 and SOX2 Using qRT-PCR

Biomarkers determined to be responsive to KDM1A inhibitors, ASCL1 and SOX2, as determined in example 2 were then verified by Taqman qRT-PCR analysis on the same set of SCLC cell lines described in examples 1 and 2, including two additional SCLC cell lines, one identified as sensitive to KDM1Ai treatment (DMS53) and the other identified as partially sensitive to KDM1Ai treatment (NCIH 526).

Gene expression analysis by qRT-PCR

Total RNA was extracted using the RNeasy Mini kit and cDNA was obtained using the High Capacity RNA-to-cDNA Master Mix (ThermoFisher Scientific #4390779) according to standard procedures. qRT-PCR was performed using LightCycler 480Probes Master (PNT-L-034; Roche #04887301001) and using pre-designed and pre-optimized TaqMan gene expression analysis from ThermoFisher Scientific. qRT-PCR was performed in triplicate using Lightcycler 480Instrument II (Roche; PNT-L-035). Cp values were analyzed by qRT-PCR in triplicate using the following Taqman primer/probe sets:

-ASCL 1: hs04187546_ g1(Life Technologies; amplicon length 81bp, targeting exon 1-2 borders, RefSeq NM-004316.3, see SEQ ID No.1)

-SOX 2: hs01053049_ s1(Life Technologies; amplicon length 91bp, targeting exon 1-1 boundary, RefSeq NM _003106.3, see SEQ ID No.3)

-GAPDH: hs02758991_ g1(Life Technologies; amplicon length 93bp, targeting exon 6-7 boundaries, Refseq NM-001256799.2)

Results

ASCL1 and SOX2 gene expression in this set of SCLC cell lines as measured by qRT-PCR are shown in table 3. The table reports the Cp values. Exp. r. experimental replicates; av.: averaging; n.d.: not detected. Each experimental replicate is the average of three technical replicates. Quantification of the same RNA for each sample was analyzed by qRT-PCR. For comparison, the average Cp expression of the GAPDH reference gene was reported, ranging from 23 to 26Cp in all samples (see table 3).

Table 3:

table 4 shows the average Cp values of all experimental replicates of ASCL1 and SOX2 in SCLC cell lines as measured by qRT-PCR. The color code displayed is based on gene expression level; the darker the color, the higher the expression of the biomarker.

Table 4:

as shown in tables 3 and 4, expression of ASCL1 or SOX2 in KDM1Ai resistant cells was not detected at all (Cp value >40), or had an absolute Cp value above 35, indicating very low expression. On the other hand, all KDM1Ai sensitive SCLC cell lines expressed ASCL1 and SOX2, and these findings were confirmed using the RNASeq data described in example 2. In partially sensitive cell lines, one was shown to express high levels of ASCL1 and SOX2, while the other had very low expression of ASCL1 and SOX 2.

These results are also graphically represented in fig. 2, which is a dot plot showing gene expression of ASCL1 (Y-axis) and SOX2 (X-axis) of KDM1A inhibition sensitive, partially sensitive or resistant SCLC cell lines as described above, measured by qRT-PCR (absolute Cp value). The plotted values are the average of independent experiments, as shown in table 3. One of the cell lines showed a Cp value of SOX2 expression above 40; this is indicated by the dots shown in brackets in fig. 2.

Thus, the results described herein further demonstrate that SCLC sensitive to KDM1A inhibitor treatment generally exhibits high expression of ASCL1 and SOX2, while resistant SCLC has low expression of one or both of ASCL1 and SOX 2. Thus, ASCL1 and SOX2 levels can be used as predictive biomarkers to identify SCLC cells or subjects with SCLC that have an increased likelihood of responding to treatment with a KDM1A inhibitor.

Example 4 evaluation of predictive biomarkers for KDM1A inhibitor response in an expanded set of SCLC cell lines

To further validate the biomarkers ASCL1 and SOX 2as biomarkers responsive to KDM1A inhibition, a larger data set was created and analyzed. SCLC viability assay data obtained using ary-1001 is integrated with publicly available data on SCLC Cell lines responsive to two additional KDM1A inhibitors (GSK2879552 and GSK-LSD1) and then used to classify a larger group of SCLC Cell lines as sensitive or resistant to KDM1A inhibitor treatment, as described in Mohammad et al, Cancer Cell 2015,28:57-69 (see in particular fig. S2A-B therein, incorporated herein by reference). An expanded set of such SCLC cell lines is shown in table 5 below. The chemical structures of GSK2879552 and GSK-LSD1 are provided in the specification.

Table 5:

wherein:

cell lines were sensitive to KDM1Ai, but required more than 4 days of treatment to observe the effect, as shown by the data for the other two KDM1A inhibitors tested over a longer period.

Some cell lines were classified as partially sensitive because the sensitivity of these cell lines to KDM1A inhibition depends on the assay, with different results having been obtained under other assay conditions. The NCIH526 cell line was classified as partially sensitive because although it was sensitive under the test conditions herein, it was reported to be resistant to ary-1001 in other assays. NCIH2081 was classified as partially sensitive because some response to ORY-1001 was observed in vitro, but the maximum growth inhibition was very low (10%).

Expression of ASCL1 and SOX2 was then assessed in these KDM1Ai sensitive and resistant cells using the Cell Line transcriptome dataset (Cancer Cell Line Encyclopedia; https:// ports. branched. Expression. org/CCLE; CCLE _ Expression _ Entrez _2012-09-29.gct. txt) as planned by the Broad study. The gene expression of ASCL1, SOX2 and GAPDH in SCLC cell lines as described in the CCLE database (Affymetrix microarray data; RMA values) is shown in Table 6 below. The p-value for the two-tailed student's t-test was calculated using Microsoft Office Excel.

Table 6:

ASCL1 and SOX2 were differentially expressed between cell lines sensitive and resistant to KDM1Ai treatment (table 6). Specifically, the mean expression of ASCL1 for the sensitive and resistant cell lines was 12.25 and 7.19 normalized probe intensity units (p-value ═ 3.0E-05), respectively. In the same cell line, the mean normalized probe intensity values for SOX2 were 8.96 (sensitive group) and 5.75 (resistant group) (p-value ═ 2.0E-03). The expression of the GAPDH reference gene was similar in all samples (sensitive cell line 14.60 average normalized probe intensity units, resistant cell line 14.56 average normalized probe intensity units, p-value 6.1E-01).

Consistent with the results described in examples 2 and 3, 7 of the 8 KDM1 Ai-sensitive SCLC cell lines highly expressed ASCL1 and SOX2, while almost all resistant SCLC cell lines expressed low levels of one or both of ASCL1 and SOX 2. This is further highlighted in fig. 3, which shows a dot plot showing gene expression of ASCL1 and SOX2 in an expanded set of SCLC cell lines sensitive, partially sensitive or resistant to KDM1A inhibition as measured by microarray Affymetrix analysis (RMA values). As shown in figure 3, a significant enrichment of cell lines sensitive to KDM1A inhibitor was observed in cell lines with high levels of ASCL1 and SOX2, and vice versa, in cell lines with low expression of one or both of ASCL1 and SOX2, a significant enrichment of cell lines resistant to KDM1A inhibitor was observed.

Thus, the results described in example 4 herein further support the use of these two biomarker combinations to identify/select subjects more likely to respond to treatment with a KDM1A inhibitor, such as ary-1001.

Sensitive and resistant cell lines were analyzed for gene expression values for each of the selected biomarkers (excluding "partially sensitive" cell lines from the analysis) using GraphPad Prism 5.01 software using a receiver operating characteristic curve (ROC curve). The ROC curve is created by plotting True Positive Rate (TPR) versus False Positive Rate (FPR) at various threshold settings. Fig. 4 shows ROC curves for ASCL1 (fig. 4A) and SOX2 (fig. 4B) expression to distinguish SCLC cells sensitive and resistant to KDM1A inhibitors. On the basis of the ROC curve for each biomarker, a threshold level for the best trade-off between selectivity and specificity (highest likelihood ratio) for each is reported. For the expression data set in Table 6, only using ASCL1 biomarkers with threshold levels ≧ 8.935, KDM1A sensitive and resistant cell lines can be distinguished with a sensitivity of 68.42% and a specificity of 100%. For a given expression dataset, using only SOX2 biomarkers with threshold levels ≧ 8.030, KDM 1A-sensitive and resistant cell lines can be distinguished with a sensitivity of 84.21% and a specificity of 87.50%.

Responses to KDM1A inhibition were then predicted in the SCLC cell line panel described in table 6 (excluding "partially sensitive" cell lines in the analysis) based on the combination of ASCL1 and SOX2 using different algorithms.

In a first example, a boolean conjunctive model classification algorithm was constructed based on biomarkers that simultaneously meet the respective thresholds for exceeding ASCL1 and SOX2 described above (fig. 4A and B); a score of "1" is obtained if the algorithm meets the conditions specified in the first row of table 7 below, and a score of "0" is obtained otherwise. Then, when the score exceeds a threshold, i.e., >0 (in this case, equal to 1), the cell line is classified as more likely to respond to KDM1 Ai; when the score is equal to 0, the classification is unlikely to respond (see table 7 below). Finally, we evaluated the performance of the boolean conjunctive classification algorithm. The sensitivity (also called true positive rate; TPR), specificity (also called true negative rate; TNR), positive predictive value (PPV, also called precision) and Negative Predictive Value (NPV) were calculated, as well as the geometric mean between TPR and TNR, PPV and NPV. Sensitivity was calculated as the ratio between true positives and the total number of positives. Similarly, specificity was obtained as the ratio between the number of true negatives and the total number of negatives. Positive predictive value (PPV; also called precision) and Negative Predictive Value (NPV) were calculated using the following formulas:

PPV＝TP/(TP+FP)

NPV＝TN/(TN+FN)

where TP represents the number of true positives, TN represents the number of true negatives, FP represents the number of false positives and FN represents the number of false negatives.

Table 7:

ASCL1/SOX2 identified highly sensitive (88%), specific (100%), accurate (100%) and negative predictive value (95%).

Alternatively, Support Vector Machine (SVM) modeling using the DTREG predictive modeling software is used to develop algorithms that can be used to classify samples using biomarkers. In particular, the combination of ASCL1 and SOX2 was evaluated as predictive biomarkers of responsiveness to KDM1A inhibitors using two different algorithms with different kernel functions, polynomials, and Radial Basis Functions (RBF) and respective parameters (C: cost parameters; γ: Kernell coefficients) as shown in table 8 (top). Using the scores and thresholds (functions) determined by these algorithms, cell lines were classified as more likely or less likely to respond to the indicated KDM1Ai, and the performance of the SVM model to classify samples is shown at the bottom of table 8, showing the specificity, sensitivity and confusion matrices that the SVM model generated was used to predict sensitivity and resistance using a combination of ASCL1 and SOX2 expression.

Table 8:

using this algorithm, with simultaneous polynomial and RBF fitting, the SVM algorithm with the ASCL1/SOX2 combination has high sensitivity (88%) and high specificity (> 95%).

Alternatively, an algorithm was developed using linear regression of the DTREG predictive modeling software to classify samples as more likely to respond to KDM1Ai or less likely to respond to KDM1Ai as a function of the combination of biomarkers selected on the data set in table 6 (table 9). The algorithm calculates a score for each sample and classifies the cell line as sensitive or resistant to KDM1Ai by comparing the score for each sample to a threshold. The performance of the linear regression model generated based on the ASCL1/SOX2 biomarker combination had 87.5% sensitivity and 94.74% specificity to predict sensitivity to KDM1Ai (table 9).

Table 9 below shows the parameters, specificity, sensitivity and confusion matrix for the generated linear regression model to predict sensitivity and resistance using a combination of ASCL1 and SOX2 expressions. C: a cost parameter; gamma is the kernel coefficient of the corresponding function.

Table 9:

overall, the use of different classification algorithms (boolean conjunctive model, SVM model and linear model) combining the expression levels of ASCL1 and SOX2 demonstrated that this dual marker signature has the best performance in predicting SCLC responsiveness to KDM1A inhibitors.

Example 5 correlation of ASCL1 and SOX2mRNA with protein expression levels by Western Blotting (WB) and fluorescence Immunohistochemistry (IF)

To test the correlation between biomarker mRNA and protein levels, ASCL1 and SOX2 were analyzed by WB in SCLC cell lines expressing these biomarkers high, medium or low/undetectable, and by fluorescent immunohistochemistry of SCLC cell pellets of the same sections and SCLC patient-derived xenografts (PDX) with known mRNA levels.

5.1 analysis of SOX2 and ASCL1 in SCLC cell lines by WB

Human Small Cell Lung Carcinoma (SCLC) cell lines NCI-H146, NCI-H510A, NCI-H446, and NCI-H526 at 37 ℃ and 5% CO₂In RPMI medium (Sigma) supplemented with 2mM glutamine and 10% fbs (Sigma). Cell pellets of 500 ten thousand exponentially growing cells were generated for extraction of whole protein in RIPA buffer (Sigma) supplemented with protease inhibitor (Sigma), followed by determination of the levels of ASCL1(Abcam, ab213151) and SOX2(Abcam, ab97959) by WB in 12% page (life technologies). Proteins were transferred using the iBlot System (Life Technologies), and after secondary antibody incubation and washing, blots were developed with ECL Prime (Amersham) and photographed with G: BOX Chemi XRQ (Syngene). Ponceau S staining of transfer blots was used as loading control. WB signals were quantified using Image J. The integrated density of each WB band was normalized by the corresponding integrated density of total protein stained with ponceau red and correlated with the NCI-H146 signal.

Results

ASCL1 and SOX2 protein levels were analyzed by WB of SCLC cell lines expressing these biomarkers in high, medium or low/undetectable (as determined by qRT-PCR (example 3) and confirmed with publicly available Affymetrix mRNA expression data from Cancer Cell Line Encyclopedia (CCLE) (see example 4)). The WBs obtained are shown in FIG. 5A, and the corresponding quantification of ASCL1 and SOX2 protein levels in the NCI-H146, NCI-H510A, NCI-H446 and NCI-H526 cell lines is shown in FIG. 5B. The protein expression levels of ASCL1 and SOX2 by WB correlated with their corresponding mRNA levels, since ASCL1 was not detected in NCI-H446, ASCL1 and SOX2 were not detected in NCI-H526 cells, and their expression was the highest in NCI-H146, confirming the expression levels of both mRNAs and the specificity of the antibody used. The correlation between protein and mrna (ccle affymetrix) levels of SOX2 and ASCL1 are plotted in fig. 6A and 6B, respectively; good correlation was observed with an R value of 0.8957 for SOX2 and 0.9910 for ASCL 1.

5.2 analysis of SOX2 and ASCL1 by fluorescence immunohistochemistry on SCLC cell pellets

1000 ten thousand exponentially growing cells (NCI-H146, NCI-H510A, NCI-H446, and NCI-H526) were fixed in 10% formalin (Sigma) for 1 hour at room temperature, washed in 1XPBS (Sigma), pelleted, contained in 1.3% agarose (Sigma), then dehydrated and contained in paraffin for microtome sectioning. The 5 μm sections were placed on Superfrost slides, dewaxed for 5 minutes in HistoChoice detergent (Sigma), performed twice, and hydrated by decreasing ethanol series (2X 100% 5 minutes, 90% 1 minute, 70% 1 minute, 30% 1 minute, 2X running water). The sections were then heat-induced antigen retrieval in boiling citrate buffer 1x (sigma) at pH 6 for 20 minutes. After 20 minutes at room temperature, the slides were washed in PBS-Triton X1000.1% (0.1% PBS-Tx) and blocked in 5% goat serum in 0.1% PBS-Tx for 1 hour at room temperature. After blocking, excess fluid was removed by capillary action with a paper towel, and the sections were incubated overnight at 4 ℃ with primary antibody diluted in 1% goat serum in 0.1% PBS-Tx (1: 500 dilution of Abcam ab97959 for SOX 2; 1:100 dilution of Abcam ab213151 for ASCL1) and the corresponding negative control (1% goat serum in 0.1% PBS-Tx only). After washing 3 times for 5 minutes each in 0.1% PBS-Tx, the slides were incubated with goat anti-rabbit Alexa Fluor 546 secondary antibody (1:1500 dilution, Life Technologies A11010) for 1 hour at room temperature in the dark. Each was washed 5 minutes in 0.1% PBS-Tx, 5 times, and after 5 washes, excess liquid was removed by capillary action with a paper towel and the samples were packaged in Fluoroshield packaging medium supplemented with dapi (sigma). DAPI is 4', 6-diamidino-2-phenylindole, a fluorescent dye that binds strongly to a-T rich regions of DNA and is used to stain cell nuclei. Nuclear signals from DAPI staining were used to identify and analyze co-localization of nuclear-specific SOX2 and ASCL1 signals (see below). Images were taken in a Zeiss Axio fluorescence microscope with a coupled AxioCam camera (Zeiss).

Quantification of IF intensity

The IF images were processed and quantified using imageJ software. For quantification of SOX2 and ASCL1 nuclear specific signals, a mask surrounding the nuclear coverage area was created from the DAPI stained image and then converted to the corresponding IF image to determine the integrated density in the selected area defined by the nuclei only.

Results

To validate the antibodies used for fluorescence immunohistochemistry and to further confirm the correlation between ASCL1 and SOX2 protein and mRNA levels, the same antibodies used for WB were tested in fluorescence immunohistochemistry to examine the levels of these biomarkers in SCLC cell pellets in formalin fixed paraffin embedded sections. The fluorescent immunohistochemistry of SOX2, ASCL1, and negative control is shown in fig. 7 (fig. 7A-SOX2, fig. 7B-ASCL1, fig. 7C-negative control using secondary antibody only (AF 546)). Consistent with the data shown previously, SOX2 levels were high in the NCI-H146 cell line, moderate in the NCI-H510A cell line, low in the NCI-H446 cell line, and no nuclear-specific expression was detected in the NCI-H526 cells. On the other hand, the ASCL1 level was high in NCI-H146, moderate in NCI-H510A, and absent/undetectable in NCI-H446 and NCI-H526 cells. No expression was observed in the negative control (AF 546: Alexa Fluor 546) using secondary antibody alone. Based on the staining intensity, the following staining intensity levels were established and used from now on in the following examples: a high level will be defined as "level 3", a medium level as "level 2", a low level as "level 1" and no signal as "level 0".

Table 10 below shows the quantification of nuclear biomarker signals in immunofluorescence images as shown in FIG. 7 and the corresponding intensity levels and RMA values from CCLE Affymetrix in NCI-H146, NCI-H510A, NCI-H446 and NCI-H526 cells. The signal was background corrected and expressed relative to the NCI-H146 signal (equivalent to 100%).

The quantification of nuclear signals for the cell lines analyzed revealed a highly statistically significant correlation between ASCL1 and SOX2 protein levels and the corresponding mRNA levels as detected by fluorescence immunohistochemistry (see fig. 8A-SOX2 and 8B-ASCL 1). The R values for SOX2 and ASCL1 were 0.8710 and 0.9594, respectively.

In view of the good correlation obtained between ASCL1 and SOX2mRNA and protein levels as shown in examples 5.1 and 5.2 above, it was demonstrated that measuring the mRNA or protein levels of ASCL1 and SOX2, ASCL1 and SOX2 can be used as predictive biomarkers in response to KDM1A inhibition.

5.3 paired by fluorescence immunohistochemistry on patient derived SCLC xenograft (PDX) Tissue Microarray (TMA) Analysis of SOX2 and ASCL1

To further confirm the correlation between SOX2 and ASCL1 mRNA and protein levels in human samples, SOX2 and ASCL1 IF were performed on patient-derived SCLC xenograft tissue microarrays with available SOX2 and ASCL1 RNASeq data.

Tissue microarray sections containing 44 patient-derived SCLC xenografts were purchased from Molecular Response (now Crown biosciences) and their corresponding available RNAseq data were downloaded from https:// OncoExpress.

TMA was dewaxed in HistoChoice detergent (Sigma) for 5 minutes, twice and hydrated by decreasing ethanol (2X 100% 5 minutes, 90% 1 minute, 70% 1 minute, 30% 1 minute, 2X running water). The sections were then heat-induced antigen retrieval in boiling citrate buffer 1x (sigma) at pH 6 for 20 minutes. After 20 minutes at room temperature, the slides were washed in PBS-Triton X1000.1% (0.1% PBS-Tx) and blocked in 5% goat serum in 0.1% PBS-Tx for 1 hour at room temperature. After blocking, excess fluid was removed by capillary action with a paper towel, and the sections were incubated overnight at 4 ℃ with primary antibody in 1% goat serum diluted in 0.1% PBS-Tx (1: 500 dilution of Abcam ab97959 for SOX 2; 1:100 dilution of Abcam ab213151 for ASCL1) and the corresponding negative control (1% goat serum contained in 0.1% PBS-Tx only). After washing for 5 minutes in 0.1% PBS-Tx and 3 times, the slides were incubated with goat anti-rabbit Alexa Fluor 546 secondary antibody (1:1500 dilution, Life Technologies A11010) for 1 hour at room temperature in the absence of light. Each was washed 5 minutes in 0.1% PBS-Tx, 5 times, and after 5 washes, excess liquid was removed by capillary action with a paper towel and the samples were packaged in Fluoroshield packaging medium supplemented with dapi (sigma). Images were taken in a Zeiss Axio fluorescence microscope with a coupled AxioCam camera (Zeiss).

Classification of PDX IF intensity

IF staining was analyzed by Zeiss Axio fluorescence microscopy and tumor regions were defined by nuclear morphology. The nuclear-specific signal intensity in the tumor region was visually divided into four levels, as described in example 5.2:

samples with expression levels above the biomarker threshold 1 for both ASCL1 and SOX2, respectively (i.e. medium to high expression levels 2 and 3), or alternatively samples with ASCL1/SOX2 boolean junctional scores above the threshold (>0) (i.e. both conditions for each biomarker are met simultaneously), are classified as originating from patients more likely to respond to KDM1Ai, based on the intensity obtained in the fluorescence immunohistochemistry performed on SCLC cell pellets and their corresponding known mRNA expression (example 5.2). In fig. 9, samples with scores exceeding the threshold are indicated as "positive", and samples with scores not exceeding the threshold are indicated as "negative".

Statistical analysis

RNAseq (Log) for SOX2 and ASCL1 was performed using GraphPad Prism software₂FPKM) and IF (visual scoring) data sets with a confidence interval of 95%.

Results

SOX2 and ASCL1 IF were performed on patient-derived SCLC xenograft tissue microarrays with available SOX2 and ASCL1 RNASeq data. All samples were analyzed visually under a fluorescence microscope for tumor area defined by nuclear morphology and intensity levels were given according to the observed nuclear signal within the tumor area, as described above. Representative ASCL1 and SOX2 staining from SCLC PDX TMA for each staining intensity classification level is shown in fig. 9.

In two independent stains of two consecutive SCLC PDX TMA sections (N35 and N43, respectively), the IF visual score and RNASeq data showed a highly statistically significant (P <0.0001) correlation, Spearman r values for SOX2 were 0.7535 and 0.7659 (fig. 10A and 10B), and for ASCL1 were 0.8803 and 0.8989 (fig. 10C and 10D), respectively.

The above results demonstrate that SOX2 and ASCL1 protein and mRNA levels are also correlated in patient-derived samples, so by measuring the mRNA or protein levels of ASCL1 and SOX2, the mRNA or protein levels of ASCL1 and SOX2 can be used as predictive biomarkers in response to KDM1A inhibition in human samples.

Example 6: ASCL1 and SOX2 can be detected in exosomes

Exosomes are microvesicles present in body fluids, the contents of which reflect the proteasome, the genome and the transcriptome of the parent cell. Exosomes therefore constitute an excellent minimally invasive tool for quantitative biomarker detection. Therefore, we tested whether our assays for predictive biomarkers of KDM1A inhibitor responsiveness, ASCL1 and SOX2, are suitable for use in methods using exosome-containing samples.

Exosomes were isolated by precipitation from SCLC cell lines and SOX2 and ASCL1 protein levels were determined by WB.

Secretion of exogenous pathogenic factorDetection of ASCL1 and SOX2 in vivo

2000 million NCI-H146, NCI-H510A, NCI-H446, and NCI-H526 SCLC cells were seeded in 20ml RPMI medium (Sigma) supplemented with 2mM glutamine (Sigma) and 10% exosome-free FBS (System biosciences) and incubated at 37 ℃ and 5% CO₂The flask was incubated in a T75 in a humid atmosphere. After 48 hours, 15ml of well resuspended cells were spun at 2.000Xg for 30 minutes at room temperature, the supernatant was transferred to a clean tube and the cell pellet was kept at-20 ℃. Exosome precipitation was performed using total exosome-separating agent (from cell culture medium) (Life Technologies) using 10ml of clarified medium, all following the manufacturer's instructions. The exosome pellet was then resuspended in 80 μ l 1xSDS loading buffer for WB or in 40 μ l RIPA buffer supplemented with protease inhibitors (Sigma) for protein extraction. After quantification, 2 volumes of RIPA extract were mixed with 1 volume of 3 xssds loading buffer, heated to 95 ℃ and kept at-20 ℃ until used for WB.

The exosome isolation method was validated by NCI-H510A cells grown in the presence of 5 μ M exosome release inhibitor GW4869 (selelcchem) or vector under the specified conditions described above, and exosomes were isolated using total exosome-isolating reagents (from cell culture medium) (Life Technologies) following the manufacturer's instructions entirely.

Cell pellets kept at-20 ℃ were used to extract proteins in RIPA buffer (Sigma) supplemented with protease inhibitors. After protein quantification using protein assay dye reagent (Bio-Rad), 7 μ g of total protein previously heated in 1XSDS loading buffer at 95 ℃ was used for WB 12% PAGE (Life technologies) with 15 μ l of exosome protein extract using ASCL1(Abcam, ab213151), SOX2(Abcam, ab97959) and antibodies to lung cancer exosome-specific CD151 marker (Abcam, ab 33315). The blot was developed with ECL Prime (Amersham) and photographed with G: BOX Chemi XRQ (Syngene). Ponceau S staining of transfer blots was used as loading control.

Results

To investigate the feasibility of performing ASCL1 and SOX2 assays in exosomes, microvesicle fractions from NCI-H146, NCI-H510A, NCI-H446 and NCI-H526 SCLC cells were isolated by precipitation and ASCL1 and SOX2 protein levels were analyzed by WB using the method described in example 5.1. The results obtained are shown in fig. 11, indicating that both ASCL1 and SOX2 can be detected in exosomes and that their expression is consistent with that in the parent cell (see fig. 5). ASCL1 was not detected in both NCI-H446 and NCI-H526-derived exosomes (see FIG. 11), but high ASCL1 was detected in NCI-H146 and NCI-H510A cells. In turn, no SOX2 was detected in the NCI-H526-derived exosomes (FIG. 11), also consistent with the expression of SOX2 reported in FIG. 5.

Furthermore, ASCL1, SOX2 and the lung cancer specific exosome marker CD151 signal were significantly reduced or eliminated in the exosome fraction of NCI-H510A cells treated with 5 μ M GW4869 (exosome release inhibitor) compared to the vector, while the expression of these proteins in parental cells remained unchanged (fig. 12). Thus, these results indicate that the exosome-derived ASCL1 and SOX2 signals are specific for this microvesicle fraction obtained by precipitation and validate the exosome isolation method employed.

Overall, the results of this example 6 demonstrate that using exosomes as starting materials/samples, measurements of protein levels of the predictive biomarkers of the invention ASCL1 and SOX2 can be performed to determine responsiveness to KDM1A inhibitors.

The present invention relates to the following nucleotide and amino acid sequences:

the sequences provided herein are available in the NCBI database and can be retrieved from www.ncbi.nlm.nih.gov/sites/entrezdb ═ gene; these sequences also relate to annotated and modified sequences. The invention also provides techniques and methods in which homologous sequences and variants of the conciseness sequences provided herein are used. Preferably, such "variants" are genetic variants, such as splice variants.

Exemplary amino acid sequences and nucleotide sequences of human ASCL1 and SOX2 are shown in SEQ ID NOS: 1 through 4 below. Exemplary nucleotide and amino acid sequences of human GAPDH (glyceraldehyde-3-phosphate dehydrogenase) used as a control gene in some embodiments are shown in SEQ ID NOS: 5 and 6.

SEQ ID No.1 nucleotide sequence, mRNA, encoding homo sapiens Achaete-Scute family bHLH transcription factor 1(ASCL1)

NCBI reference sequence: NM _ 004316.3. The coding region ranges from nucleotide 572 to nucleotide 1282 (highlighted in bold). It will be understood that the mRNA corresponds to (i.e. is identical to) the following sequence, except that the "t" (thymidine) residue is replaced by a "uracil" (u) residue.

Source

SEQ ID No.2 amino acid sequence and protein of Chiense Achaete-Scute family bHLH transcription factor 1(ASCL1)

UniProtKB/Swiss-Prot ASCL1_ human, P50553

SEQ ID No.3 nucleotide sequence encoding homo sapiens SRY-box2(SOX2), mRNA

NCBI reference sequence: NM _ 003106.3. The coding region ranges from nucleotide 438 to nucleotide 1391 (highlighted in bold). It will be understood that the mRNA corresponds to (i.e. is identical to) the following sequence, except that the "t" (thymidine) residue is replaced by a "uracil" (u) residue.

Source

SEQ ID No.4 amino acid sequence of homo sapiens SRY-box2(SOX2), protein

UniProtKB/Swiss-Prot:SOX2_HUMAN,P48431

SEQ ID No.5 homo sapiens glyceraldehyde-3-phosphate dehydrogenase (GAPDH), mRNA

NCBI reference sequence: NM _ 002046.6. The coding region ranges from nucleotide 77 to nucleotide 1084 (highlighted in bold). It will be understood that the mRNA corresponds to (i.e. is identical to) the following sequence, except that the "t" (thymidine) residue is replaced by a "uracil" (u) residue.

Source

SEQ ID No.6 amino acid sequence of homo sapiens glyceraldehyde-3-phosphate dehydrogenase (GAPDH), protein UniProtKB/Swiss-Prot: G3P _ HUMAN, P04406

Claims

1. A method of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from a patient prior to initiating treatment comprising a KDM1A inhibitor.

2. The method of claim 1, further comprising identifying the patient as more likely to respond to a treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold.

3. The method of claim 1, further comprising using the levels of ASCL1 and SOX2 in the sample to generate a score for the sample, wherein the patient is identified as more likely to respond to a treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

4. A method of identifying an SCLC patient who may benefit from treatment comprising a KDM1A inhibitor, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to commencing the treatment comprising a KDM1A inhibitor.

5. The method of claim 4, further comprising identifying the patient as likely to benefit from treatment comprising a KDM1A inhibitor when the level of each of ASCL1 and SOX2 in the sample exceeds a threshold.

6. The method of claim 4, further comprising using the levels of ASCL1 and SOX2 in a sample to generate a score for the sample, wherein the patient is identified as likely to benefit from treatment comprising a KDM1A inhibitor when the score in the sample exceeds a threshold.

7. A method of selecting a treatment for an SCLC patient, the method comprising measuring the levels of ASCL1 and SOX2 in a sample from the patient prior to initiating the treatment.

8. The method of any of claims 1 to 7, wherein the ASCL1 levels and SOX2 levels are mRNA expression levels.

9. The method of claim 8, wherein the mRNA expression level is measured by qRT-PCR.

10. The method of any of claims 1 to 7, wherein the ASCL1 levels and SOX2 levels are protein expression levels.

11. The method of claim 10, wherein the protein expression level is measured by fluorescence immunohistochemistry.

12. The method of any of claims 1 to 11, wherein the sample is a biopsy sample.

13. The method of any one of claims 1 to 12, further comprising recommending, prescribing, or administering to the patient a therapeutically effective amount of treatment comprising a KDM1A inhibitor if the patient is identified as more likely to respond to treatment comprising a KDM1A inhibitor.

Use of a KDM1A inhibitor for the treatment of a SCLC patient, wherein the patient has been identified as more likely to respond to treatment comprising a KDM1A inhibitor using the method according to any of claims 1 to 12 prior to initiating treatment comprising a KDM1A inhibitor.

15. A method of treating an SCLC patient, the method comprising administering to the patient a therapeutically effective amount of a treatment comprising a KDM1A inhibitor if the patient has been identified as more likely to respond to treatment comprising a KDM1A inhibitor using the method of any of claims 1-12 prior to initiating treatment comprising a KDM1A inhibitor.

Use of ASCL1 and SOX2 in a method of identifying SCLC patients more likely to respond to treatment comprising a KDM1A inhibitor.

17. The method of claims 1-13, KDM1A inhibitor for use according to claim 14, the method of treatment according to claim 15 or the use according to claim 16, wherein the KDM1A inhibitor is (trans) -N1- ((1R,2S) -2-phenylcyclopropyl) cyclohexane-1, 4-diamine or a pharmaceutically acceptable salt thereof.

18. The method according to claims 1 to 13 or 17, KDM1A inhibitor for use according to claims 14 or 17, the method of treatment according to claims 15 or 17 or the use according to claims 16 or 17, wherein the patient is a human patient.

19. A kit for assessing the likelihood of an SCLC patient responding to a treatment comprising a KDM1A inhibitor, the kit comprising one or more reagents for measuring the level of ASCL1 and SOX2 in a sample, and optionally, instructions for use.